Early detection of HIV-positive status and entry into treatment is the most efficacious way to prevent the spread and/or progression of the disease. However surveillance data and cohort studies indicate Latinos in the United States (U.S.) are more likely than non-Latino blacks and whites to delay testing, present to care with an AIDS defining illness, and die within one year of learning their HIV-positive status. Recent studies have shown that U.S. immigrants from Mexico and Central America are more likely to delay testing and treatment compared to other foreign-born groups; and Latinos in Southeastern settlement areas may delay seeking care longer than Latinos in more established immigrant destinations. Because time lived in the U.S. and acculturation are often associated with increased behavioral risk and suboptimal health outcomes among Latinos, it is important to understand how combinations of social influences may affect HIV-risk, socio-medical trust, and likelihood of HIV-testing over time. The proposed research will apply Simons-Morton's Social Influences Model to test theoretical links between social influences, trust in medical providers, and likelihood of HIV testing among heterosexually active immigrant men and women. The proposed research uses data from three NIH and extramurally funded studies (2007 - 2009) to address three specific aims: (1) Assess validity of the WFU Physician Trust Scale among Spanish-speaking men and women from Mexico and Central America; (2) Estimate the structural relationships between sojourner social support, acculturative factors, and physician trust adjusting for length of residence in the U.S. and (3) Examine whether socio-medical trust moderates the HIV risk and testing relationships among men in the HoMBReS-2 HIV prevention intervention. The proposed research will advance theory and measurement of psychosocial processes in the most rapidly growing U.S. populations experiencing the most severe AIDS-related outcomes.